Online reputation management (ORM) is the practice of monitoring, influencing and repairing what people and AI systems find about a person or organization online. It covers search results, review sites, news coverage and AI-generated answers. The work combines listening, publishing accurate content, responding to reviews and requesting removals where a platform's rules allow.
Last updated: November 14, 2026
What is the short version of this guide?
Online reputation management comes down to knowing what is said about a brand, correcting what is wrong, and publishing enough accurate material that the correct version is what people and AI systems find first.
- ORM covers four places: search results, review sites, news and social coverage, and AI-generated answers.
- A Harvard Business School study found that a one-star increase in Yelp rating led to a 5 to 9 percent increase in revenue for independent restaurants.
- The Federal Trade Commission's fake review rule took effect on October 21, 2024, so buying reviews is not a reputation strategy.
- Google's Results about you tool can remove some personal contact details from search results, but it does not remove most negative coverage.
- Muck Rack's May 2026 report found that earned media supplied about 84 percent of the links AI systems cited, so third-party coverage shapes how AI describes a brand.
What is in this online reputation management guide?
This guide covers what ORM includes, why it matters, a five-step process, how to handle reviews and removals, and how AI answers change the work.
- What ORM includes
- Why it matters
- The five-step process
- Which fix fits which problem
- Handling negative reviews
- What can be removed
- ORM and AI answers
- ORM, PR and crisis communications
- Cost of ORM
- Frequently asked questions
What does online reputation management include?
Online reputation management includes five activities: monitoring, review management, content publishing, removal and correction requests, and measurement. Each activity targets a different place where a reputation forms.
| Activity | What it covers | Example action |
|---|---|---|
| Monitoring | Search results for the brand and its leaders, review sites, news, social posts and AI answers | Search the brand name in a private window each week and log the first ten results |
| Review management | Requesting, answering and flagging reviews on Google, Yelp and industry sites | Reply to each review with specifics and flag only reviews that break the platform's rules |
| Content publishing | Owned pages and independent coverage that state accurate facts | Publish a leadership page and a fact sheet, then earn coverage from outside sources |
| Removal and correction | Requests to platforms, publishers and search engines | File a Results about you request when a home address appears in search |
| Measurement | Changes in what appears and how it is described | Rerun the same searches and AI prompts every month and compare the results |
A person can run a small version of this alone. A company with many locations, executives or products usually assigns an owner and a written process, because the same facts have to stay consistent across many listings.
Why does online reputation management matter?
Online reputation management matters because buyers use visible ratings and search results to judge a business they cannot inspect directly. The best-documented evidence comes from a study of restaurant reviews.
Harvard Business School's Michael Luca combined Yelp ratings with Washington State Department of Revenue data for Seattle restaurants between 2003 and 2009. In his paper, Reviews, Reputation, and Revenue: The Case of Yelp.com, he found that a one-star increase in Yelp rating led to a 5 to 9 percent increase in revenue. The effect came from independent restaurants. Ratings had no measurable effect on chain restaurants, whose brand already told customers what to expect.
The study covers one industry in one state, so the 5 to 9 percent figure should not be applied to other businesses. Its mechanism does carry over. When a buyer has little other information, a visible rating or a top search result stands in for the reputation the business has not yet had time to build.
The same logic now applies to AI assistants. A buyer who asks ChatGPT, Gemini or Perplexity about a company receives a written description, and that description is assembled from pages the assistant retrieves. Reputation work therefore has to cover those pages as well as the review sites.
How does online reputation management work step by step?
Online reputation management works in five steps: audit, triage, fix, publish and monitor. The order matters because the audit decides which fixes are worth the effort.
- Audit search results. Search the brand, its founders and its main products in a private browser window, record the first page of results, and label each result as owned, earned, a review or hostile.
- Audit reviews and AI answers. Read the newest reviews on each platform, then ask ChatGPT, Gemini, Perplexity and Google AI Overviews the questions a buyer would ask, and record what each one says and which sources it cites.
- Triage by cause. Sort every problem into false information, outdated information, accurate but unflattering information, or missing information, because each type needs a different fix.
- Fix and publish. Request removal or correction of false items, update outdated ones, answer accurate criticism in public, and publish pages that state the correct facts in plain language.
- Monitor on a schedule. Repeat the same searches and prompts every month and compare them with the baseline from step one.
Triage is the step most teams skip. A team that treats every negative result as a removal problem spends months on requests that platforms will reject, while the real gap, such as a missing company fact sheet, stays open.
Which fix fits which type of problem?
Each type of reputation problem has a usual fix, and using the wrong one wastes time or creates a second problem. The table pairs the four problem types, plus personal data, with the fix that usually applies.
| Problem type | Example | Usual fix | What to avoid |
|---|---|---|---|
| False information | A review from someone who was never a customer | Flag it to the platform with evidence that it breaks the rules, and ask an attorney about defamation | Posting replies from fake accounts |
| Outdated information | A directory listing with an old address | Ask the site to update it and correct every owned profile | Ignoring it, since assistants repeat old facts |
| Accurate but unflattering | A news story about a past recall | Publish a factual update on what changed and earn newer coverage | Threatening the publisher |
| Missing information | No page that explains what the company does | Publish a fact sheet, leadership pages and a plain description of services | Leaving the description to third-party sites |
| Personal data | A home address on a data broker site | File a Results about you request and ask the broker to remove the listing | Assuming Google removal deletes the page |
The accurate but unflattering row is the one most often handled badly. Honest criticism cannot be removed, but it can be answered and gradually outweighed by newer, accurate material. The guide to online reputation repair walks through that sequence.
How should a business handle negative reviews?
A business should answer negative reviews in public, state the facts once, and flag only the reviews that break the platform's content rules. A calm, specific reply is read by future customers as much as by the reviewer.
- Thank the reviewer and name the specific problem they raised.
- State what was done or will be done in one or two sentences.
- Offer an offline channel for details and never share private customer information in the reply.
- Flag a review only when it breaks the platform's rules, such as a review from someone with no connection to the business.
- Ask satisfied customers for reviews through a consistent process, without payment and without asking only for positive ratings.
The law limits what a business can do. The Federal Trade Commission announced its final rule on consumer reviews and testimonials on August 14, 2024, and it took effect on October 21, 2024. According to a Mondaq summary of the rule, it bars writing or buying fake reviews, including AI-generated ones, and allows civil penalties of up to $51,744 per violation, a figure the government adjusts for inflation. The rule also covers buying positive or negative reviews, undisclosed insider reviews and using threats to suppress honest reviews.
A reputation program built on fake reviews therefore carries legal risk as well as the risk that a platform or a journalist exposes it.
What can and cannot be removed from search results?
Search engines remove a narrow set of content, mainly personal contact details, certain sensitive identifiers and content that breaks a law or policy, while accurate negative news coverage generally stays in results.
Google's Results about you tool, launched in 2022, lets people find search results that show their phone number, home address or email address and request removal. Google's help page for the tool says Google may approve those requests and notes that other personally identifiable information can also be submitted. Removing a result from Google hides the page from Search only. The page remains on its own site, so the stronger fix is to ask the site owner to take it down.
Content that a court has found defamatory can go through Google's legal removal process. Most other negative coverage cannot be removed and can only be outranked by accurate, newer material. An attorney should assess any defamation claim before a brand sends a legal demand, because a weak demand can draw more attention to the original item.
How does online reputation management apply to AI answers?
Online reputation management applies to AI answers because assistants describe a brand using the sources they retrieve, so reputation work now includes the third-party pages those assistants cite.
Muck Rack's May 2026 What Is AI Reading? report analyzed more than 25 million links cited by ChatGPT, Claude and Gemini across 17 industries. It found that earned media supplied about 84 percent of citations and journalism about 27 percent. Press releases made up 1.1 percent and paid or advertorial content 0.3 percent. Muck Rack counts academic, government, encyclopedic and third-party corporate sources as earned media alongside journalism, so the 84 percent figure is broader than news coverage. The figures change with each edition of the report.
The practical mechanism is repetition across independent sources. When several credible sites state the same fact about a brand, a retrieval system is more likely to return that fact. One outdated article on a high-authority site can keep appearing in answers long after the brand has changed.
The guide to tracking brand visibility in AI search sets out a fixed-prompt routine for this audit, and the post on what search shows after a PR crisis explains how the written record is formed. For category-level evidence, see the AI citations research.
How does ORM differ from PR and crisis communications?
Online reputation management is an ongoing practice focused on what appears online, while public relations builds coverage and crisis communications responds to a specific incident. The three overlap, and a brand often needs all of them.
PR earns the coverage that gives a reputation its raw material. Crisis communications handles the first days of an incident, when statements and spokespeople matter most. ORM begins after the first response, when the record of the incident has to be corrected, updated and balanced with newer material. The post on how AI is transforming reputation management covers the monitoring side in more detail.
How much does online reputation management cost?
Online reputation management cost depends on scope. A one-time cleanup of a few listings costs far less than a continuing program covering search results, reviews, earned coverage and AI answers.
A small business with a handful of problems can often run monitoring and review replies in-house. Larger brands with many locations, executives or past incidents usually need a team. 5W's online reputation management retainers run $15,000 to $75,000 per month, and the audit sets the scope. The reputation management services page describes how that work is organized.
How can you run a 30-minute reputation check today?
Open a private browser window and search the brand name, the founder's name and the brand name plus the words reviews, scam and complaints. Record the first ten results for each search. Open the three most recent reviews on each major review platform. Ask one AI assistant three buyer questions about the brand and write down the sources it cites. Mark each item as false, outdated, accurate but unflattering or missing. That list is the baseline for every later fix.
Frequently asked questions
What is online reputation management in simple terms?
Online reputation management is the work of finding out what the internet says about a person or company, correcting what is wrong and publishing accurate material so the correct version is easy to find. It covers search results, reviews, news and AI answers.
Does ORM remove negative news articles?
Online reputation management rarely removes accurate news articles. Publishers decide what to keep, and search engines generally leave accurate coverage in results. The usual approach is to publish newer, accurate material and to ask for corrections where a story contains a factual error.
Can negative reviews be removed?
Negative reviews can be removed only when they break the platform's content rules, for example when the reviewer has no connection to the business or the review contains prohibited content. An honest negative review generally stays, and the best response is a public, specific reply.
Is it legal to pay for positive reviews?
No. The Federal Trade Commission's rule on consumer reviews, effective October 21, 2024, prohibits buying positive or negative reviews and writing or buying fake reviews, and it allows civil penalties against violators.
How long does online reputation management take?
No fixed timeline applies. Removal requests depend on each platform's review process, and outranking an old page depends on how much newer, accurate material exists. A monthly comparison against the baseline shows whether the work is moving results.
Does online reputation management include AI search?
Yes. AI assistants describe brands using retrieved sources, so a reputation program should record what each assistant says and which pages it cites, then correct the sources that are wrong or outdated.
What should a brand do first?
A brand should run the 30-minute check above this week and sort every item it finds into false, outdated, accurate but unflattering or missing. Fix the false and outdated items first, because those are the ones platforms and publishers will act on, and publish a plain fact sheet for anything that is missing.
